Published in the October/November 2010 issue of Neural Networksis a study done by a team of researchers from the University of Washington's Institute for Learning and Brain Sciences which shows that babies can be tricked into believing robots are sentient beings. In the study, the researchers took a sample of 64 18-month-old babies and individually tested them by having the babies sit on their parents' laps, facing a humanoid robot named Morphy, remote-controlled by an unseen researcher. Another researcher, Rechele Brooks, sitting next to Morphy, would then engage in a 90-second skit in which Brooks would interact with the robot, asking questions like “Where is your tummy?” and “Where is your head?” Morphy would respond accordingly by pointing to its different parts, and the robot would also imitate a few arm movements like waving back and forth. After the skit, the Brooks would leave the room, leaving the baby and the robot alone. Morphy would then beep and shift slightly to get the baby's attention, and then the robot would turn to look at a nearby toy. In 13 out of 16 cases, the baby would follow the robot's gaze as if what the robot looks at might be of interest to the baby. [1] At 18 months, babies are able to distinguish between a swiveling chair and a person's movement, and the baby will only follow the person. This study suggests that the baby has decided that the robot is a human being. [2]
In contrast, in only 3 of 16 cases would the baby interact with the robot when they were left alone with the robot without first seeing Morphy engage in human interactions. Naturally, babies look to adults for guidance on how to interpret things, but this research also suggests that the robot's human-like appearance is less important than its behavior when it comes to establishing trust. [1] (Or this means that a robotic takeover is imminent should the steely machines ever take this study into consideration. Watch for your young.)
Overview
Published in the October/November 2010 issue of Neural Networks is a study done by a team of researchers from the University of Washington's Institute for Learning and Brain Sciences which shows that babies can be tricked into believing robots are sentient beings. In the study, the researchers took a sample of 64 18-month-old babies and individually tested them by having the babies sit on their parents' laps, facing a humanoid robot named Morphy, remote-controlled by an unseen researcher. Another researcher, Rechele Brooks, sitting next to Morphy, would then engage in a 90-second skit in which Brooks would interact with the robot, asking questions like “Where is your tummy?” and “Where is your head?” Morphy would respond accordingly by pointing to its different parts, and the robot would also imitate a few arm movements like waving back and forth. After the skit, the Brooks would leave the room, leaving the baby and the robot alone. Morphy would then beep and shift slightly to get the baby's attention, and then the robot would turn to look at a nearby toy. In 13 out of 16 cases, the baby would follow the robot's gaze as if what the robot looks at might be of interest to the baby. [1] At 18 months, babies are able to distinguish between a swiveling chair and a person's movement, and the baby will only follow the person. This study suggests that the baby has decided that the robot is a human being. [2]
In contrast, in only 3 of 16 cases would the baby interact with the robot when they were left alone with the robot without first seeing Morphy engage in human interactions. Naturally, babies look to adults for guidance on how to interpret things, but this research also suggests that the robot's human-like appearance is less important than its behavior when it comes to establishing trust. [1] (Or this means that a robotic takeover is imminent should the steely machines ever take this study into consideration. Watch for your young.)
[1] http://machineslikeus.com/news/babies-treat-social-robots-sentient-beings
[2] http://www.popsci.com/science/article/2010-10/babies-recognize-robots-human-through-social-interaction (It is strongly advised that everyone click on this link, if any, just to see the goofy picture of two of the researchers that articles tend to use – it looks like it came straight from the Sears Portrait Studio.)
History
(There is no text here yet.)
Opinion
(There is no text here yet.)
Future Trends?
(There is no text here yet.)